{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:P6RSPFMERNUF27RASPREWLRUOX","short_pith_number":"pith:P6RSPFME","schema_version":"1.0","canonical_sha256":"7fa32795848b685d7e2093e24b2e3475dda4968472b5d8530467f44505bbf8b8","source":{"kind":"arxiv","id":"2608.00267","version":1},"attestation_state":"computed","paper":{"title":"LoopsBench: From Harness Engineering to Loop Engineering in Benchmarking Coding Agent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Dayi Lin, Dongmei Zhang, Han Li, He Ye, Jiaheng Liu, Pengfei Gao, Qingwei Lin, Rili Feng, Saravan Rajmohan, Yingqi Zhao, Zhemin Fang","submitted_at":"2026-07-31T20:18:25Z","abstract_excerpt":"Coding agent infrastructure is shifting from harness engineering toward loop engineering as coding agents are deployed for sustained long-horizon software development. Existing benchmarks often center on localized tasks or end-state outcomes, offering limited insight into sustained execution. We introduce LOOPSBENCH, a long-horizon benchmark for loop engineering in coding agent evaluation. Each task is a dependency DAG over separately testable development units with source-evidenced prerequisite edges. LOOPSBENCH comprises 112 tasks from authentic sources spanning 8 programming languages and 9"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2608.00267","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-07-31T20:18:25Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f40f57c8a92a72cac66067855c3bc4952d3f9b843fd130adcee6d0a073d8ae9d","abstract_canon_sha256":"54426179c27a85402a5d533e9e6a8a3bc9e3446e86efa7744c28772f0aa11fa6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T00:34:42.680197Z","signature_b64":"MuxoF+RfKA/XS7f+1t/eTk7GWSofzqsa0Zc3lqT1l7PZQDOteHRmrtCw1IG3dIKsPOoTj3CBd092dB4bo31/DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7fa32795848b685d7e2093e24b2e3475dda4968472b5d8530467f44505bbf8b8","last_reissued_at":"2026-08-04T00:34:42.678476Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T00:34:42.678476Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LoopsBench: From Harness Engineering to Loop Engineering in Benchmarking Coding Agent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Dayi Lin, Dongmei Zhang, Han Li, He Ye, Jiaheng Liu, Pengfei Gao, Qingwei Lin, Rili Feng, Saravan Rajmohan, Yingqi Zhao, Zhemin Fang","submitted_at":"2026-07-31T20:18:25Z","abstract_excerpt":"Coding agent infrastructure is shifting from harness engineering toward loop engineering as coding agents are deployed for sustained long-horizon software development. Existing benchmarks often center on localized tasks or end-state outcomes, offering limited insight into sustained execution. We introduce LOOPSBENCH, a long-horizon benchmark for loop engineering in coding agent evaluation. Each task is a dependency DAG over separately testable development units with source-evidenced prerequisite edges. LOOPSBENCH comprises 112 tasks from authentic sources spanning 8 programming languages and 9"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00267","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2608.00267/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2608.00267","created_at":"2026-08-04T00:34:42.679671+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.00267v1","created_at":"2026-08-04T00:34:42.679671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00267","created_at":"2026-08-04T00:34:42.679671+00:00"},{"alias_kind":"pith_short_12","alias_value":"P6RSPFMERNUF","created_at":"2026-08-04T00:34:42.679671+00:00"},{"alias_kind":"pith_short_16","alias_value":"P6RSPFMERNUF27RA","created_at":"2026-08-04T00:34:42.679671+00:00"},{"alias_kind":"pith_short_8","alias_value":"P6RSPFME","created_at":"2026-08-04T00:34:42.679671+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX","json":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX.json","graph_json":"https://pith.science/api/pith-number/P6RSPFMERNUF27RASPREWLRUOX/graph.json","events_json":"https://pith.science/api/pith-number/P6RSPFMERNUF27RASPREWLRUOX/events.json","paper":"https://pith.science/paper/P6RSPFME"},"agent_actions":{"view_html":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX","download_json":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX.json","view_paper":"https://pith.science/paper/P6RSPFME","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.00267&json=true","fetch_graph":"https://pith.science/api/pith-number/P6RSPFMERNUF27RASPREWLRUOX/graph.json","fetch_events":"https://pith.science/api/pith-number/P6RSPFMERNUF27RASPREWLRUOX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX/action/storage_attestation","attest_author":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX/action/author_attestation","sign_citation":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX/action/citation_signature","submit_replication":"https://pith.science/pith/P6RSPFMERNUF27RASPREWLRUOX/action/replication_record"}},"created_at":"2026-08-04T00:34:42.679671+00:00","updated_at":"2026-08-04T00:34:42.679671+00:00"}